Annu. Rev. Anthropol. 1992.21:491-516

Copyright © 1992 by Annual Reviews 1nc. All rights reserved

SOCIAL ANTHROPOLOGY OF

TECHNOLOGY

Bryan Pfaffenberger

Division of Humanities, School of Engineering and Applied Science, University of Virginia, Charlottesville, Virginia 22901

KEYWORDS: activity systems, technological change, sociotechnical systems, ritual, artifacts

At the onset of the 20th century, anthropologists such as Balfour, Marett, and Haddon could readily identify three spheres of strength in anthropological research: material culture, social organization, and physical anthropology (49). The study of technology and material culture, however, was about to be jettisoned, and with stunning finality. By 1914, Wissler (103:447) complained

that the study of these subjects "has been quite out of fashion." Researchers were giving their attention to "language, art, ceremonies, and social organiza­tion" in place of the former almost obsessive concentration on the minute description of techniques and artifacts, and on the tendency to study artifacts without rcgard for their social and cultural context. As I aim to show in this chapter, the anthropological study of technology and material culture is poised, finally, for a comeback, if in a different guise. Its findings may signifi­cantly alter the way anthropologists analyze everyday life, cultural reproduc­tion, and human evolution.

If this all-but-forgotten field is to play such a role, it must overcome nearly a century of peripheral status. In anthropology's quest for professionalism, material-culture studies came to stand for all that was academically embarrass­ing: extreme and conjectural forms of diffusionist and evolutionist explana­tion, armchair anthropology, "field work" undertaken by amateurs on collecting holidays, and the simplistic interpretation of artifacts shorn of their social and cultural context. Malinowski, for instance, condemned the "purely

technological enthusiasms" of material culture ethnologists and adopted an "intransigent position" that the study of "technology alone" is "scientifically sterile" (69:460). The study of technology and material culture, a topic that

491 0084-6570/92/1015-0491$02.00

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

492 PFAFFENBERGER

was (and is still) perceived as "dry, even intellectually arid and boring" (92, 5),

was relegated to museums, where--{)ut of contact with developments in social

anthropology and deprived of ethnographic experience-museum scholars

lacked the resources to advance the field. For their part, cultural andu-opolo­

gists argued that studying techniques and artifacts could only deflect anthro­pologists from their proper role-that is, from studying culture. Kroeber & Kluckhohn, for example, dismissed the term material culture out of hand,

arguing that "what is culture is the idea behind the artifact" (55 :65). "Accord­

ingly," Kroeber argued, "we may forget about this distinction between mate­

rial and nonmaterial culture, except as a literal difference, that is sometimes of

practical convenience to observe" (54:296). For anthropology, jettisoning ma­

terial culture studies was a necessary step in establishing the scientific basis, the intellectual appeal, and the distinctive subject matter of the discipline.

Periodic attempts have been made to revive the seriously ill patient (e.g. 21,

34, 49 ,5 9 ,65 ,66 ,8 5, and 88), with their pace quickening in the 1980s (43, 44,

62,6 3, 72, 7 3, 8 4, 92, 96). Yet, arguably, no real resuscitation has taken place, owing largely to the continued insouciance with which Anglo-American an­

drropologists regard the study of material culture and technology. In a recent restatement of the Kroeber & Kluckhohn view, Bouquet (13:352) condemns

the "recent bids to reinstate the 'materiality' of material culture," as if such

bids stemmed from some Philistine conspiracy, against which she prefers rccognition of the "hegemony of linguistic approaches to the object world."

Noting that the excesses of early material-culture scholars were rightly pillo­

ried, S illitoe laments that the "mud seems to have stuck more to artifacts and their study ... than to the [evolutionists'] wild-guess theories," which have

themselves enjoyed a modest comeback (92, p. 6). As it stands, a topic with

which anthropology was once closely identified-the cross-cultural study of

technology and material culture-has been largely taken up by scholars work­

ing in other fields, such as the history of technology and the interdisciplinary

field known as science and technology studies (STS), or by andrropologists with marginal appointments in museums or in the general studies divisions of engineering and technical colleges.

Despite the peripheral status of the anthropology of technology and mate­rial culture, compelling questions remain: What is technology? Is technology a

human universal? What is the relationship between technological development and cultural evolution? Are there common themes in the appropriation of

artifacts that bridge capitalist and precapitalist societies? How do people em­

ploy artifacts to accomplish social purposes in the course of everyday life? What kind of cultural meaning is embodied in technological artifacts? How does culture influence technological innovation-and how does technological

innovation influence culture? These questions are far from trivial-and, arguably, only anthropology can

answer them. No other discipline offers sufficient comparative depth or appro­priate methodologies. The challenge still remains, as Malinowski himself put

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 493

it, to understand the role of technology as "an indispensable means of ap­proach to economic and sociological activities and to what might be called native science" (69:460). And the challenge is even greater now that scholars generally concede that language, tool use, and social behavior evolved in a process of complex mutual interaction and feedback. Summing up the consen­sus of a conference titled "Tools, Language, and Intelligence: Evolutionary Implications," Gibson concludes that "We need to know more about the ways in which speaking, tool-using, and sociality are interwoven into the texture of everyday life in contemporary human groups" (29:263).

In this chapter, I argue that social anthropology has already discovered a great deal about human technological activity-especially when anthropologi­cal findings are interpreted in the context of recent, stunning advances in the sociology of scientific knowledge ( 1 1), the history and sociology of technol­ogy, and the emergent field known as science and technology studies (STS). Collectively, these fields, without much anthropological involvement, have developed a concept, the sociotechnical system concept (48), that refuses to deny the sociality of human technological activity. Developed mainly in social and historical studies of industrial societies, the sociotechnical system concept, I seek to show, serves fruitfully to integrate anthropological findings about preindustrial societies into a coherent picture of the universals of human technology and material culture. The central objective of this review, then, is to convey the sociotechnical system concept to an anthropological audience, and to show how it resolves key controversies within anthropology. The results should prove of interest to anthropologists working in fields as diverse as cultural ecology, ritual, symbolic anthropology, ethnoarchaeology, archae­ology, and human evolution studies.

One reason for the rapid advance of STS is its refusal to accept the myths of science and technology at face value. Mulkay (74), for example, shows that sociology's refusal to develop a sociological analysis of scientific knowledge stems from sociologists' uncritical acceptance of a mythic Standard View of science. I suggest that the achievement of a truly social anthropology of technology likewise requires extending anthropology's recent productive ven­ture into reflexivity (18, 70)-specifically, by making the mythic Standard View of technology explicit, and resolutely questioning its implications. For this reason, this essay begins with the Standard View of technology, and although its purpose is to present the sociotechnical system concept and ex­plore its implications for anthropology, it is organized as a series of attacks on the implications of the Standard View.

THE Sf ANDARD VIEW OF TECHNOLOGY

Like the Standard View of science (74: 19-21), the Standard View of technol­ogy underlies much scholarly as well as popular thinking. A master narrative of modern culture, the Standard View of technology could be elicited, more or

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

494 PFAFFENBERGER

less intact, from any undergraduate class. Occasionally, it is made explicit in anthropological writings (e.g. 39). By suggesting that such a Standard View exists, J do not mean to imply that every scholar who has advanced some part of it necessarily endorses the rest. In what follows, I deliberately use the masculine pronoun to stand for humankind; to do otherwise would strip the Standard View of its gender ideology.

Necessity is the mother of invention. As Man has been faced with severe survival challenges, certain extraordinary individuals have seen, often in a brilliant flash of inspiration, how to address the challenge of Need by applying the forces, potentialities, and affordances of Nature to the fabrication of tools and material artifacts. The power of Nature is there, waiting to be harnessed, to the extent that the inventor can clear away the cobwebs of culture to see the world from a purely utilitarian standpoint. In this we see Man's thirst for Progress.

Form follows function. To be sure, Man decorates his tools and artifacts, but artifacts are adopted to the extent that their form shows a clear and rational relationship to the artifacts' intended function-that is, its ability to satisfy the

need that was the raison d'etre of the artifact's creation. Thus, a society's material culture becomes a physical record of its characteristic survival adap­tation; material culture is the primary means by which society effects its repro­duction. The meaning of human artifacts is a surface matter of style, of surface burnish or minor symbolization.

By viewing the material record of Man's technological achievements, one can directly perceive the challenges Man faced in the past, and how he met these challenges. This record shows a unilinear progression over time, because tech­nology is cumulative. Each new level of penetration into Nature's secrets builds on the previous one, producing ever more powerful inventions. The digging stick had to precede the plough. Those inventions that significantly increase Man's reach bring about revolutionary changes in social organization and subsistence. Accordingly, the ages of Man can be expressed in terms of technological stages, such as the Stone Age, the Iron Age, the Bronze Age, and so on. Our age is the Information Age, brought on by the invention of the computer. Overall, the movement is from very simple tools to very complex machines. It was also a movement from primitive sensorimotor skills (techniques) to highly elaborate systems of objective, linguistically encoded knowledge about Nature and its potential (technology).

Now, we live in a material world. The result of the explosion of technological knowledge has been a massive expansion of Man's reach, but with lamentable and unavoidable social, environmental, and cultural consequences: We live in a fabricated environment, mediated by machines. Technology was more authentic when we used tools, because we could control them. Machines, in contrast, control us. Thus one can identify a Great Divide or Rupture when Man lost his authenticity as a cultural creature, his Faustian depth as a being living in a world of cultural meaning, and gave himself over to a world ruled by instrumentalism and superficiality. This Rupture was the Industrial Revolution, which launched the Age of the Machine. As the primacy offunction over aesthetics rips through culture, we increasingly live in a homogenous world offunctionally driven design coherence. Our culture has become an inauthentic one in which reified images of technology predominate. We can define ourselves only by purchasing plastic,

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 495

ersatz artifacts made far away. To retain some measure of authenticity the young must be brought into direct contact with the great works of art and literature.

The Standard View of technology appears to be a pillar of Modernism, a cultural, literary, and artistic period noted for its extreme ambivalence toward technology. According to most scholars (e.g. 40), Modernism reached its apex between the two World Wars. In essence, Modernism represents a struggle to find a stable ground of being within the promise and peril of science and technological development. Like Siva in Hindu iconography, technology is seen through the Modernist lens as both creator and destroyer, an agent both of

future promise and of culture's destruction. Echoed perfectly in the Standard View of technology, Modernism amounts to

an extraordinary compound of the futurist and the nihilistic, the revolutionary and the conservative, the romantic and the classical. It was the celebration of a technological age and a condemnation of it; an excited acceptance of the belief that the old regimes of culture were over, and a deep despairing in the face of that fear (15:46).

Modernism is an almost unavoidable response, as Bradbury & McFarlane put it, to the "scenario of our chaos" (15:27). Accurdingly, any attempt to grasp the role of human technological activity must begin by questioning the Stand­ard View's assumptions, which could, if left unexamined, color anthropologi­cal thought.

"NECESSITY IS THE MOTHER OF INVENTION"

The Standard View puts forth a commonsense view of technology and mate­rial culture that accords perfectly with our everyday understanding. All around us are artifacts originally developed to fulfill a specific need-juicers, word processors, vacuum cleaners, and telephones; and apart from artifacts that are decorative or symbolic, the most useful artifacts-the ones that increase our fitness or efficiency in dealing with everyday life-are associated each with a specific Master Function, given by the physical or technological properties of the object itself. Extending this commonsense view one quickly arrives at a theory of technological evolution (parodied by 4:6): People need water, "so they dig wells, dam rivers and streams, and develop hydraulic technology. They need shelter and defense, so they build houses, forts, cities, and military machines .... They need to move through the environment with ease, so they invent ships, chariots, charts, carriages, bicycles, automobiles, airplanes, and spacecraft."

The New Archaeology, which sought to put archaeology on a firm modern­ist footing, puts forward a view of technology and artifacts firmly in accord with the Standard View and its presumption of need-driven technological evolution. Culture, according to Binford (7), is an "extrasomatic means of adaptation"; thus technology and material culture form the primary means by

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

496 PFAFFENBERGER

which people establish their viability, given the constraints imposed upon

them by their environment and the demands of social integration. It follows, as Binford argued in 19 65 (8), that every artifact has two dimensions, the pri­

mary, referring to the instrumental dimension related to the artifact's function, and the secondary, related to the artifact's social meaning and symbolism. Echoing this view, Dunnell makes explicit the connection that is assumed between an artifact's function and group survival: The artifact's function is that which "directly enhances the Darwinian fitness of the populations in which they occur" (23: 19 9 ). Style, in contrast, is something added on the surface, a burnish or decoration, that might play some useful role in symboliz­ing group solidarity but is decidedly secondary. In the Modernist view, there are universal human needs, and for each of these there is an ideal artifact. For

the primitive technologist, discovering such an artifact is like discovering America: It was there before the explorers finally found it-and to the extent that anyone bothers to look, it will be found, and inevitably adopted (although it might be resisted for a time). The tale of Man's rise, then, is the story of increasing technological prowess, as digging sticks develop into ploughs,

drums into telephones, carts into cars. The Standard View of technology offers a seemingly "hard" or "tough­

minded" view of artifacts and technological evolution, but there is ample evidence that its "hardness" dissolves when examined critically. What seems to us an incontrovertible need, for which there is an ideal artifact, may well be generated by our own culture's fixations. Basalla (4:7- 1 1) demonstrates this point forcefully with respect to the wheel. First used for ceremonial purposes in the Near East, the wheel took on military applications before finally finding

transport applications. In Mesoamerica, the wheel was never adopted for trans­port functions, given the constraints of terrain and the lack of draught animals. Even in the Near East, where the wheel was first invented, it was gradually given up in favor of camels. Basalla comments, "A bias for the wheel led Western scholars to underrate the utility of pack animals and overestimate the contribution made by wheeled vehicles in the years before the camel replaced the wheel" (4: 1 1). Against all Modernist bias, Basalla's vicw echoes the findings of recent social anthropologists who have argued that it is impossible to identify a class of "authentic" artifacts that directly and rationally address "real" needs (2, 22:7 2; 8 7). Culture, not nature, defines necessity. One could reassert that a "hard" or "tough-minded" approach requires the recognition, after all, that people must eat, and so on, but it is abundantly evident that a huge variety of techniques and artifacts can be chosen to accomplish any given utilitarian objective (9 1).

The supposed functions of artifacts, then, do not provide a clear portrait of a human culture's needs (38 ), and what is more, one call110t unambiguously infer from them precisely which challenges a human population has faced. The natives of chilly Tierra del Fuego, after all, were content to do without cloth­ing. Accordingly, some archaeologists and social anthropologists would break

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 497

radically with the Standard View in asserting that material culture does not play a decisive role in shaping a human group's adaptation to its environment. Golson (32) notes that a basic stone toolkit survives intact through "revolu­tionary" changes in subsistence in both the classic Old World sites and in the New Guinea highlands. A survey of the New Guinea tools, Golson concludes, "revealed none that is indispensable to any form, from the simplest to the most complex, of Highlands agricultural practice, except the stone axe or adze and the digging stick which are not only common to all but also serviceable in other than agricultural contexts" (32:161). Summarizing the evidence from social anthropology, Sahlins (86:81) puts this point well: "For the greater part of human history, labor has been more significant than tools, the intelligent efforts of the producer more significant than his simple equipment." Sahlins'

view is echoed by Lemonnier (62:151), who notes that the "search for corre­spondences between technical level and 'stage' of economic organization does not seem likely to lead to a theory of the relation between technical systems and society, other than one so over-simplified and general that it quickly loses all interest." Material culture alone provides only a shadowy picture of human adaptations.

If techniques and artifacts are not the linchpins of human adaptation, as is so often surmised, then radical redefinitions are in order. It is not mere tech­nology, but technology in concert with the social coordination of labor, that constitutes a human population's adaptation to its environment. In most prein­dustrial societies, technology plays second fiddle to the human capacity to invent and deploy fabulously complex and variable social arrangements. How, then, should we define technology? Spier (93:2), for instance, defines technol­ogy as the means by which "man seeks to modify or control his natural environment." This definition is clearly unsatisfactory. It assumes, a priori, that Man's inherent aim is domination or control of nature; and, anyway, it is wrong, since (as has just been argued) techniques and artifacts are secondary to the social coordination of labor in shaping human adaptations. One could broaden the definition of technology to include the social dimension. But because the term "technology" so easily conjures up "merely technical" activ­ity shorn of its social context (77), I believe it preferable to employ two definitions, the one more restricted, and the other more inclusive. Technique

(following 62, 63) refers to the system of material resources, tools, operational sequences and skills, verbal and nonverbal knowledge, and specific modes of work coordination that come into play in the fabrication of material artifacts. Sociotechnical system, in contrast, refers to the distinctive technological activ­ity that stems from the linkage of techniques and material culture to the social coordination of labor. The proper and indispensable subjects of a social an­thropology of technology, therefore, include all three: techniques, sociotechni­cal systems, and material culture.

The sociotechnical system concept stems from the work of Thomas Hughes on the rise of modem electrical power systems (45, 46; for applications of the

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

498 PFAFFENBERGER

concept, see 68, 81). According to Hughes, those who seek to develop new technologies must concern themselves not only with techniques and artifacts;

they must also engineer the social, economic, legal, scientific, and political context of the technology. A successful technological innovation occurs only when all the elements of the system, the social as well as the technological, have been modified so that they work together effectively. Hughes (45) shows how Edison sought to supply electric lighting at a price competitive with natural gas (economic), to obtain the support of key politicians (political), to cut down the cost of transmitting power (technical), and to find a bulb filament of sufficiently high resistance (scientific). In a successful sociotechnical sys­tem, such as the electric lighting industry founded by Edison, the "web is seamless": "the social is indissolubly linked with the technological and the economic" (60: 112). In short, sociotechnical systems are heterogeneous con­structs that stem from the successful modification of social and nonsocial actors so that they work together harmoniously-that is, so that they resist dissociation (60: 166-l7)-i.e. resist dissolving or failing in the face of the system's adversaries. One or more sociotechnical systems may be found in a given human society, each devoted to a productive goal.

Extending Hughes's concept, Law (60) and Latour (57) emphasize the difficulty of creating a system capable of resisting dissociation. A system builder is faced with natural and social adversaries, each of which must be controlled and modified if the system is to work. Some of them are more obdurate, and some of them more malleable, than others. In illustrating this point, Law shows that the sociotechnical system concept applies fruitfully to the study of preindustrial technology, in this case the rise of the Portuguese mixed-rigged vessels in the 14th and early 15th centuries. The real achieve­ment, argues Law, was not merely the creation of the mixed-rigged vessel, with its increased cargo capacity and storm stability. Equally important was the magnetic compass, which allowed a consistent heading in the absence of clear skies; the simplification of the astrolabe, such that even semieducated mariners could determine their latitude; exploration that was specifically in­tended to produce tables of data, against which position could be judged; and an understanding of Atlantic trade winds, which allowed ships to go forth in one season and come back in another. To achieve the necessary integration of all these factors, the system builders had to get mariners, ship builders, kings, merchants, winds, sails, wood, instruments, and measurements to work to­gether harmoniously. The system they created resisted dissociation; they were able to sail out beyond the Pillars of Hercules, down the coast of Africa, and soon around the globe.

Although it is no easy trick to construct a system resistant to dissociation, sociotechnical systems are not inevitable responses to immutable constraints; they do not provide the only way to get the job done. People unfamiliar with technology usually gravely understate the degrees of latitude and choice open to innovators as they seek to solve technical problems (48). More commonly,

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 499

one sees a range of options, each with its tradeoffs, and it is far from obvious

which, if any, is superior. In virtually every technical area, there is substantial latitude for choice. For instance, Lemonnier (62) points to the apparently

arbitrary variation of techniques as one moves across the New Guinea high­

lands; such variation is to be found, Lemonnier notes, even among those

"functional" (as opposed to "stylistic") aspects of a tool that are directly implicated in its action upon material (62: 160). It would be wrong to attribute

a system's "success" (i.e. in resisting dissociation) to the choice of the "cor­

rect" technique or social-coordination method.

By analogy to the sociology of scientific knowledge (11), this point can be

formulated as a principle of symmetry. In the sociology of scientific knowl­

edge, this principle countered an older sociology of science that explained the

success of a theory by its conformity to the Truth, while ascribing the failure of

another theory to social factors (bias, influence, "interests," etc). The principle

of symmetry calls for precisely the same kind of social explanation to be used in accounting for the success as well as the failure of a theory-or, by exten­

sion, of a sociotechnical system. Accordingly, it would violate the principle of

symmetry to argue that one system succeeds because its builders chose the

"right" techniques, the ones that really "work." Of apparently successful sys­tems, we can say unly that the system builders have apparently succeeded in bringing to life one out of a range of possible systems that could achieve its

goal (e.g. trapping wild pigs, growing rice, or sailing down the coast of

Africa). Such a system could be viewed as an adaptation, in line with cultural

ecology, but only by abandoning the post hoc, ergo propter hoc fallacy of

functionalism. That a sociotechnical system develops does not imply that it is

the logical system, or the only possible system, that could havc dcveloped

under the circumstances; social choice, tactics, alternative techniques, and the social redefinition of needs and aspirations all play a role in the rise of

sociotechnical systems.

An additional example, south Indian temple irrigation, should help to clar­

ify the sociotechnical system concept and its implications. A marked charac­

teristic of agriculture in medieval south India was the royal donation of wastelands to communities of Brahmans, who in turn were encouraged to

organize and supervise agricultural production. They did so by investing the lands in newly constructed temples, which provided a locus of managerial

control for the construction, maintenance, and management of complex irriga­

tion systems (42, 67) that successfully resisted drought and led to a majestic

efflorescence of south Indian Hindu culture. The heterogeneous quality of such a system is immediately evident. The system linked into a cohesive, successful system actors such as kings, canal-digging techniques, dams, flow­

ing water, modes of coordinating labor for rice production, agricultural rituals, deities, notions of social rank and authority, conceptions of mcrit flowing from donations, conceptions of caste relations and occupations, conceptions of so­

cially differentiated space, religious notions of the salutary effect of temples

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

500 PFAFFENBERGER

on the fertility of the earth, economic relations (land entitlements), trade,

temple architecture, and knowledge of astrological and astronomic cycles

(used to coordinate agricultural activities). A human sociotechnical system

links a fabulous diversity of social and nonsocial actors into a seamless web

(47). Any sociotechnical system shows the imprint of the context from which it

arose, since system builders must draw on existing social and cultural re­

sources. But it is important to stress that every sociotechnical system is in

principle a de novo construct; to make the system work, system builders draw

from existing resources but modify them to make them function within the

system. In this sense, sociotechnical-system building is almost inevitably so­

ciogenic (56): Society is the result of sociotechnical-system building. The

distinctive social formation of medieval south India, for instance, is in almost

every instance attributable to the achievement of the sociotechnical system of

temple irrigation. The system of temple irrigation draws on old ideas of gods,

kings, water, dams, castes, gifts, and all the rest, but it transforms every one of

these ideas in important ways. In this sense, the sociotechnical system concept is in accord with the structuration theory of Giddens (30): People construct

their social world using the social resources and structures at hand, but their

activities modify the structures even as they are reproduced.

A sociotechnical system, then, is one of the chief means by which humans

produce their social world. Yet sociotechnical systems are all but invisible

through the lenses provided by Western economic, political, and social theory,

as Lansing (56) discovered in his study of Balinese irrigation. From the stand­point of Western theory, irrigation is organized either by the despotic state, as

Wittfogel argued, or by autonomous village communities, as anthropologists

argued in reply. Invisible within this discourse, Lansing found, was the Bali­

nese water temple, a key component in a regional sociotechnical system

devoted to the coordination of irrigation. Lansing discovered that the rites in

these Balinese water temples define the rights and responsibilities of subsidi­

ary shrines (and with them, the subaks, or local rice-growing collectivities, that

line the watershed) through offerings and libations of holy water. By symboli­

cally embedding each local group's quest for water within the supra-local

compass of temple ritual, water temples encourage the cooperation necessary

to ensure not only the equitable distribution of water but also the regulated

flow of inundation and fallowing that proves vital for pest control and fertility.

Tellingly, the solidarity that is created is not political; the king has obvious

interests in promoting this kind of solidarity but does not actively intervene

within it. And neither is this solidarity purely economic; it crosscuts other

arenas of economic integration. A sociotechnical system engenders a distinc­

tive form of social solidarity that is neither economic nor political (47); that is

why it took so long for these systems to be "discovered" by anthropologists

indoctrinated with classical social theory.

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 501

Sociotechnical systems have remained equally invisible through the lens provided by the Standard View of technology, which refuses to deal with the

ritual dimension of technical activity. According to the Standard View, and to virtually every anthropological definition of technology, a technique is an

effective act (62:154, citing 71), as opposed to magic or religion. Spier makes this commonsense assumption explicit in excluding from "technology" any "magico-religious means" by which people seek to control nature (93:2). Such a view forestalls any consideration of the crucial role that ritual institutions

play in the coordination of labor and the network's legitimation (24, 35, 83,

95, ), a point that should already be apparent from the south Indian and Balinese examples already discussed. Among the Montagnards of highland

Vietnam (19), agriculture is no mere matter of material culture and manual labor. On the contrary, ritual is a key component of agricultural work; the rites call forth social groups to engage in specific activities, and they provide a metacommentary on the entire productive process. Sociotechnical systems may very well include ritual components with explicit productive goals that

we find "false," such as enhancing the fertility of the earth; but to ignore them is to miss the crucial role they play in the coordination of labor. I would

therefore argue that the social anthropology of technology, against all common

sense, should adopt a principle of absolute impartiality with respect to whether a given activity "works" (i.e. is "technical") or "doesn't work" (i.e. is "magico­religious"); only if we adopt such impartiality do the social dimensions of sociotechnical activity come to the fore (80).

The labor-coordination role of ritual is surprisingly widespread, and for

good reason: Ritual works surpassingly well to coordinate labor under condi­tions of statelessness or local autonomy. Among the Piaroa of lowland south America, for example, shamanic rituals employ scarce mystical knowledge to

transfer mystical powers of fertility and increase to people who feel them­selves in need of such powers; Granero views such rituals as an "essential part

of the productive practices of Piaroa society" (35 :665, Granero' s emphasis). Given their access to what Granero tellingly calls the "mystical means of reproduction," shamans legitimately claim the right to solicit and coordinate agricultural labor, as well as organize trade (1986). Under stateless or locally autonomous conditions, rituals provide the ideal medium for the coordination of labor in that they virtually rule out dissent (9): "you cannot argue with a song." In Sri Lanka, 19th-century civil servants meticulously recorded the rites of the threshing floor, which required economically significant transactions to be conducted with a superstitious scrupulousness of detail, and a special, virtually incomprehensible language. This ritual language required partici­pants to adopt an "odd shibboleth," as one observer termed it, for these vital

economic transactions; the limited vocabulary sharply constrained what could be said (79). Thus another key feature of sociotechnical systems is their

silence, the relatively insignificant role played by human language as against nonverbal communication in ritual (28) as a coordinator of technical activities.

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

502 PFAFFENBERGER

Here we see yet another reason for the invisibility of such systems within the compass of West em social theory, which excessively privileges language over nonverbal cognition and behavior (10).

A successful sociotechnical system achieves a stable integration of social and nonsocial actors, but it is no static thing: Keeping the network functioning requires constant vigilance, and it may also require additional technical or social modification. Every sociotechnical system must cope with what Hughes calls reverse salients, areas of obduracy or resistance that prevent the system from expanding or threatcn it with dissociation. On reaching the Indies, the Portuguese found that Muslims had monopolized the trade with Hindu princes; the Portuguese response was to work a good deal to make the cannon lighter and more powerful (60: 127-28). Sociotechnical systems also betray a charac­teristic life cycle (46) as they grow from invention, small-scale innovation, growth and development, and a climax of maximum elaboration and scope, followed by senescence and decay, until the system disappears or is replaced by a competing system. Such life cycles may be visible in the myriad cycles of innovation, growth, efflorescence, and decay that characterize the archaeologi­cal record.

The sociotechnical system concept, in sum, suggests that mere necessity is by no means the mother of invention, just as production alone is by no means the sole rationale for the astonishing linkages that occur in sociotechnical systems (cf 5). To be sure, sociotechnical-system builders react to perceived needs, as their culture defines them. But we see in their activities the essen­tially creative spirit that underlies sociogenesis, which is surely among the supreme modes of human cultural expression. Basalla (4: 14) puts this point well: A human technology is a "material manifestation of the various ways men and women throughout time have chosen to define and pursue existence. Seen in this light, the history of technology is part of the much broader history of human aspirations, and the plethora of made things are a product of human minds replete with fantasies, longings, wants, and desires." Basalla's point suggests that no account of technology can be complete that does not consider fully the meaning of sociotechnical activities, and in particular, the nonproduc­tive roles of technical activities in the ongoing, pragmatic constitution of human polities and subjective selves. Sociotechnical systems can be under­stood, as I argue in the next section, only by acknowledging that they produce power and meaning as well as goods.

"THE MEANING OF AN ARTIFACT IS A SURFACE

MATTER OF STYLE"

The commonsense Modernism of the Standard View desocializes human tech­nological activity, as has just been argued, by reducing the creativity of so­ciotechnical-system building to the doctrine of Necessity. In precisely the same way, the Standard View desocializes the meaning of technological arti-

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 503

facts by reducing this meaning to the artifact's alleged function, with a residual and secondary role left for the relatively superficial matter (it is claimed) of style. To recapture the sociality of human artifacts, it is necessary to turn this distinction upside down. I begin, therefore, by arguing that the supposedly "hard" part of the artifact, its function, is in reality the "softest," the one that is most subject to cultural definition.

Archaeologists commonly distinguish function and style, as has already been noted. But as Shanks & Tilley argue,

It is impossible to separate out style and the function [for instance] in either vessel shape or projectile point morphology. There is no way in which we can meaning­fully measure and determine what proportion of a vessel's shape performs some utilitarian end, the remainder being assigned to the domain of style. To take a chair-what proportion of this is functional as opposed to stylistic? No answer can be given; the style inheres in the function and vice versa. Furthermore, ascribing any specific or strictly delimited function to an object is in many, if not all cases, an extremely dubious exercise. A chair may be to sit on, it nominally fulfills this function, but chairs can also be used for standing on, or for knocking people over the head with, as pendulums, rulers, or almost anything else. This is not to deny the banal point that objects have uses and may normally be used in just one way, but it is to suggest that such a position represents, at best, a starting point rather than an end point for archaeological analysis (9 1 :92).

The views of Shanks & Tilley are echoed by Norman (75:9), who calls attention to an artifact's affordances. An affordance is a perceived property of an artifact that suggests how it should be used. Affordances are inherently multiple: Differing perceptions lead to different uses. You can drink water from a cup to quench thirst, but you can also use a cup to show you are well bred, to emphasize your taste in choosing decor, or to hold model airplane parts. But is not such a point just so much strained, special pleading? Everyone knows that chairs are primarily for sitting in; despite "minor" variations asso­ciated with specific historical styles and tastes, isn't the chair's function the pre-eminent matter? Such a distinction between function and style is common sense only to the extent that we ignore a key component of technology, ritual. In the preceding section I stressed ritual's prominent role in coordinating labor in sociotechnical systems. Here, I emphasize the equally prominent role of ritual in defining the function of material culture.

To illustrate this point with a convenient and simple example, I draw on the work of K. L. Ames on Victorian hallway furnishings (1). Ames notes that the hallway was the only space in the Victorian house likely to be used by both masters and servants. Masters and visitors of the masters' class would pass through the hall, while servants and tradesmen would be asked to sit there and wait. Ames calls attention to the contradictory character of these artifacts: They had to be visually appealing to the master class as they passed through the hall; but if they included seats, they had to be austere, without upholstery, and uncomfortable, befitting the lower social status of the messenger boys,

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

5� PFAFFENBERGER

book agents, census personnel, and soap-sellers who were made to wait there.

Plain and uncomfortable, the bench echoed the design of servants' furnishings,

which resembled (in the words of a servant quoted by Ames) the furnishings of

a penal colony. With such constant reminders of their status, the servants

would have no occasion to compare their status favorably with that of their

master and mistress. Peers and people of higher status, Ames notes, were shown past the bench and directly into the house. In short, the Victorian hallway is a special space devoted to the enactment of entry rituals.

As the Victorian hallway bench suggests, style and function cannot be distinguished as easily as the S tandard View would claim. What appears in a

naive analysis to be the superficial matter of "style" (the bench's austerity)

turns out, thanks to Ames' deeper contextual reading of hallway artifacts, to be the very "function" of the artifact (to remind servants of their status)! Note that

here the function of the artifact (to be attractive to masters and remind servants of their station) can be known only by comprehending the perceived social

role that the artifact is designed to fulfill; this perceived social role, in turn, can

be known only from a contextual analysis that fully explores the dimensions of

Victorian class sensibilities. I do not mean that the flatness and discomfort of the Victorian hallway bench were intended merely to "reflect" Victorian class

sensibilities. When employed in a ritual context, the bench was obviously intended to construct Victorian statuses in ways not obvious outside the ritual context. With this analysis in view, one can argue that the dimension of an

artifact identified by archaeologists, historians, and collectors as "style" once formed part of a now lost ritual system, and for that reason now stands out oddly and mysteriously against the artifact's supposed "function." In short, the

distinction between "function" and "style" is a product of the decontextualiza­

tion and dehistoricization of artifacts (see 43: 107-20 for an excellent illustra­

tion of this point).

Daniel Miller's work among south Indian potters (72) demonstrates that artifacts play key roles inframing ritual activities-that is, in providing cues that establish the cultural significance of the events taking place. In a little-un­derstood process that is unconscious and nonverbal, frames-though incon­

spicuous-play an important social role, establishing the context within which social action takes on meaning. For Miller artifacts are on the one hand

extremely visible and omnipresent; yet on the other hand, they operate silently and invisibly (73:109). As many anthropologists have discovered, people find

it difficult and pointless to talk about the meaning of artifacts: When pressed, informants resort to their last-ditch tactic, "Our ancestors did it this way" (62: 165). Once again we meet a familiar theme: the silence of human techno­

logical activity and its invisibility within the compass of theories that assign excessive privilege to speech and writing.

Miller's work among south Indian potters shows the cross-cultural rele­vance of my point about Victorian hallway artifacts-namely, that the "style"

of an artifact, when restored fully to its cultural context, turns out to be its

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 505

"function." But what is even more important, Miller's work suggests that this

"function" of artifacts may inspire artifact diversity, a key feature of human

technology. When many versions of an artifact are available, they can play

many roles in social life. Miller concludes: "Technology could be analyzed as

the systematic exploitation of the range of methods used in order to produce

patterned variation (72:201, my emphasis). Pushing this point further, one can argue that a major rationale for the creation of sociotechnical systems, beyond

mere Necessity, is the elaboration of the material symbols that are indispensa­

ble for the conduct of everyday life. And one can identify here another form of

linkage, as yet unexplored: the linkage between the rituals that coordinate

labor and the rituals that frame human social behavior by employing material

artifacts as cues. It seems likely that such linkages amount to a formidable

apparatus of domination, even under conditions of statelessness, thus belying

the mythos of egalitarianism in stateless societies. If no form of domination goes unresisted, then one would expect artifacts to

be employed in redressive rituals that are specifically designed to mute or

counter the invidious status implications of the dominant ritual system. I

therefore see the social use of artifacts, paraphrasing Richard Brown (12:129),

as a process of nonverbal communication. In this process, each new act of

ritual framing is a statement in an ongoing dialogue of ritual statements and

counterstatements. In the counterstatements, people whose status is adversely

affected by rituals try to obtain or modify valued artifacts, in an attempt to

blunt or subvert the dominant rituals' implications. These statements, and their

subsequent counterstatements, help to constitute social relations as a polity. I

therefore call attention to redressive technological activities, which are inter­

pretive responses to technological domination, to highlight the political dimen­

sion of technology. I call this polity-building process a technological drama.

A technological drama (78, 82) is a discourse of technological "statements"

and "counterstatements" in which there are three recognizable processes: tech­

nological regularization, technological adjustment, and technological recon­

stitution. A technological drama begins with technological regularization. In this process, a design constituency creates, appropriates, or modifies a techno­

logical production process, artifact, user activity, or system in such a way that some of its technical features embody a political aim-that is, an intention to

alter the allocation of power, prestige, or wealth (57). Because a sociotechnical

system is so closely embedded in ritual and mythic narrative, the technological

processes or objects that embody these aims can easily be cloaked in myths of

unusual power. Ford's assembly line, for example, was cloaked in the myth

that it was the most efficient method of assembling automobiles-a myth indeed, since Norwegian and Swedish experiments have shown that team

assembly and worker empowerment are just as efficient. The myth masked a

political aim: Ford saw the rigid and repetitive work roles as a way of domesti­

cating and controlling the potentially chaotic and disruptive workforce of

Southern and Eastern European immigrants (94: 153). The stratifying role of

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

506 PFAFFENBERGER

the Victorian hallway bench, to cite another example, was cloaked in a myth of

hygiene, which ascribed its plainness to its function in seating those who had

recently sojourned in the filthy streets ( 1 , 27).

Like texts, the technological processes and artifacts generated by techno­

logical regularization are subject to multiple interpretations, in which the

dominating discourse may be challenged tacitly or openly. I call such chal­lenges technological adjustment or technological reconstitution. In techno­

logical adjustment, impact constituencies-the people who lose when a new

production process or artifact is introduced---engage in strategies to compen­

sate the loss of self-esteem, social prestige, and social power caused by the

technology. In this process they make use of contradictions, ambiguities, and

inconsistencies within the hegemonic frame of meaning as they try to validate

their actions. They try to control and alter the discourse that affects them so

invidiously, and they try to alter the discursively regulated social contexts that

regularization creates. Police whose movements are tracked by surveillance

systems, for example, become adept at finding bridges and hills that break the

surveillance system's tracking signal. They can then grab a burger or chat with

another cop without having their location logged. Adjustment strategies in­clude appropriation, in which the impact constituency tries to gain access to a

process or artifact from which it has been excluded (e.g. 17). Before the

personal computer, computer enthusiasts and hobbyists learned how to hack their way into mainframe systems-;-as did the youthful Bill Gates (now the

CEO of Microsoft Corporation), who was reputed to have hacked his way into

systems widely thought to be impregnable. In technological reconstitution, impact constituencies try to reverse the implications of a technology through a symbolic inversion process I call antisignification. Reconstitution can lead to

the fabrication of counterartifacts (e.g. 51), such as the personal computer or

"appropriate technology," which embody features believed to negate or re­

verse the political implications of the dominant system.

Following Victor Turner (97:91�94, 98:32), I choose the metaphor of "drama" to describe these processes. A technological drama's statements and

counterstatements draw upon a culture's root paradigms, its axioms about social life; in consequence, technological activities bring entrenched moral

imperatives into prominence. To create the personal computer, for example,

was not only to create new production processes and artifacts, but also to bring computational power to the People, to deal the Establishment a blow by

appropriating its military-derived tools, and to restore the political autonomy

of the household vis-a-vis the Corporation. Here we see the dimension of

desire that Basalla (4) emphasizes : To construct a sociotechnical system is not merely to engage in some creative or productive activity. It is to bring to life a

deeply desired vision of social life, often with a degree of fervor that can only be termed millenarian.

In any explanation of the motivations underlying sociotechnical-system

building and artifact appropriation the role of such activities in the subjective

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 507

processes of self-definition deserves emphasis (22). In the grip of what Miller

calls the mass culture critique, we tend to treat contemporary acts of artifact

appropriation in capitalist society "as so tainted, superficial, and trite that they

could not possibly be worth investigating." Materialistc people, in addition,

are seen as "superficial and deluded, and are unable to comprehend their

position" (73:166). Yet, as Miller stresses (73:86-108) there are good grounds

for arguing that artifacts play a key role cross-culturally in the formation of the

self: Artifact manipulation and play, for example, provide the conceptual

groundwork for the later acquisition of language (100). We learn early, argues

Miller (73:215), that artifacts play key roles in a "process of social self-crea­

tion" in which artifacts are "directly constitutive of our understanding of

ourselves and others." In this sense contemporary societies, despite the rise of

the Consumer Culture, possess much in common with preindustrial societies

(2, c.f. 14:228-9): Artifacts are multiplied, elaborated, appropriated, and em­ployed in framing activities as a form of self-knowledge and self-definition, a

contention supported by the dizzying and unfathomable array of spectacular

artifacts now collecting dust in ethnological museums. Miller' s point leads

directly to a consideration of a third contention of the Standard View, the

doctrine that technological evolution has proceeded from simple to complex, and has deprived modern Man of his authenticity.

"A UNILINEAR PROGRESSION . . . FROM SIMPLE TOOLS

TO COMPLEX MACHINES"

It would be idiotic to deny that contemporary humans know a great deal more

about technology than did our predecessors. History shows cumulative trends

in virtually every field of technological endeavor. But the sociotechnical sys­

tem concept leads to the equally inescapable conclusion that an enormous

amount of human knowledge about building sociotechnical systems has been utterly and irretrievably lost. I argue here that the extent of this loss can be

appreciated only by understanding the heterogeneous nature of sociotechnical systems and by radically questioning the Standard View's assumption that the

evolution of technology may be described as the shift from Tool to Machine.

Such an analysis will raise equally radical questions about the Standard

View' s notion of Rupture.

In a preindustrial society, people do not often talk about the technical

knowledge they possess. In studying weavers in Ghana, for instance, Goody was surprised by the insignificant role of questioning and answering in the

teaching of apprentices (33). Although highly elaborate systems of ethnobo­tanical classification may play key roles in subsistence systems, an enormous

amount of technological knowledge is learned, stored, and transmitted by

experiential learning, visuaVspatial thinking, and analogical reasoning. Bloch

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

508 PFAFFENBERGER

(10:187) describes the nonlinguistic learning that takes place, a form of learn­

ing very incompletely understood in the cognitive sciences: . . .

Imagine a Malagasy shifting cultivator with a fairly clear, yet supple mental model, perhaps we could say a script, stored in long-term memory, of what a 'good

swidden' is like; and that this model is partly visual, partly analytical (though not necessarily in a sentential logical way), partly welded to a series of procedures about what you should do to make and maintain a swidden. This Malagasy is going through the forest with a friend who says to him, 'Look over there at that bit of forest, that would make a good swidden.' What happens then is that, after a rapid conceptualization of the bit of forest, the model of 'the good swidden' is mentally matched with the conceptualized area of forest, and then a new but related model, 'this particular place as a potential swidden,' is established and stored in long-term memory.

Bloch argues that the linguistically derived theory of human cognition is insufficient because it cannot account for the speed with which we perform

daily tasks such as identifying a 'good swidden.' It carmot account, as Miller notes (73:102), for our ability to absorb almost instantly the social implica­tions of a furnished interior "consisting of a combination which is not only almost certainly in some degree unique, but some of whose basic elements

may also be new to us." As we use technology for practical and social pur­poses, then, we draw on a nonverbal form of human cognition whose capabili­ties clearly form an enormous, but heretofore little recognized, component of

our species' everyday intelligence. The portion of technical knowledge that

people can verbalize represents only the tip of the iceberg. The notion that technology is applied science-that it represents the practi­

cal use of logically-formulated, linguistically-encoded knowledge-is very misleading. A sociotechnical system is much better described as an activity

system, a domain of purposive, goal-oriented action in which knowledge and

behavior are reciprocally constituted by social, individual, and material phe­nomena (64, 102). As Janet and Charles Keller have emphasized, and as Bloch' s example so tellingly illustrates, an activity system constantly fluxes between being and becoming: "Action has an emergent quality which results from the continual feedback from external events to internal representations and from the internal representations back to enactment" (52: 2). Crucial to this process is an equally flexible cycling among alternative cognitive modes, including visual/spatial thinking and linguistic/classificatory thinking (53). Visual/spatial thinking is widespread in all technological activity systems, including today's high technology. (25, 26, 101, 99). But visual/spatial think­ing is silent. Competent producers and users rarely mention it. This kind of

knowledge is lost, sometimes irretrievably, in the wake of technological "pro­gress." Recreation of a system that has been lost is virtually impossible. We have no idea how some preindustrial artifacts were made, let alone how highly effective activity systems were so successfully coordinated under preindustrial conditions.

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 509

When one views a sociotechnical system as a complex heterogeneous link­

age of knowledge, ritual, artifacts, techniques, and activity, it is apparent that

much more than visual/spatial knowledge about manufacture can be lost when

a system dissociates. A human sociotechnical system involves the coordina­

tion of a massively complex network; in the case of Portuguese naval expan­

sion this network consisted of such entities as kings and queens, ships, crews,

winds, cannons, maps, sails, astrolabes, Muslims, and gold. Viewed as an

activity system, a sociotechnical system must include all the conceptual, vis­

ual, experiential, tactile, and intuitive knowledge necessary to modify these

diverse elements so that they work together harmoniously. Even in the most

"primitive" sociotechnical systems, such as those of contemporary hunters and

gatherers living in marginal environments (e.g. 61), the scope of knowledge

integration involved is phenomenal. The complexity of any human sociotech­

nical system is belied by the simplicity of its tools (32).

All human sociotechnical systems, whether "primitive" or "preindustrial," are enormously complex and inherently heterogeneous. Through recognition

of this fact one can begin a critique of the notion of Rupture that figures so prominently in the Standard View. According to the Standard view, tool use is

authentic and fosters autonomy; one owns and controls one's own tools and

isn't dependent on or exploited by others. When we use machines, in contrast, we must work at rhythms not of our own making, and we become ensnared in

the supralocal relations necessary for their production, distribution, and main­

tenance. To the extent that we become dependent on machines we do not own,

the stage is set for exploitation. We become divorced from nature, and our

conceptions of the world become pathological, through a process called reifi­

cation (a malady frequently asserted to occur only in industrial societies).

According to the doctrine of Rupture, reification occurs because we employ objects as a means of knowing ourselves. When these objects are no longer our

own authentic products, as is the case with industrially produced artifacts, our

attention is deflected from critical self-awareness to the incompletely under­

stood Other who generates the artifact (73 :44).

The concept of sociotechnical systems enables us to see to what degree the doctrine of Rupture overstates the consequences of the transition from Tool to

Machine. Although one would be foolish to deny the significant consequences

of the machine' s rise, preindustrial sociotechnical systems were themselves complex and exploitive-frequently more so than the Standard View acknow­

ledges. A preindustrial sociotechnical system unifies material, ritual, and so­cial resources in a comprehensive strategy for societal reproduction. In the

course of participation in such a system, many if not most individuals find themselves playing dependent and exploited roles. By no means is reification

restricted to industrial technology. As Lansing notes for Bali,

Water temples establish connections between productive groups and the compo­nents of the natural landscape that they seek to control. The natural world surrounding each village is not a wilderness but an engineered landscape of rice

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

5 1 0 PFAFFENBERGER

terraces, gardens, and aqueducts created by the coordinated labor of generations. Anthropomorphic deities evoke this residual human presence in an engineered landscape . . . . Each wier is the origin of an irrigation system, which has both physical and social components. The concept of the deity of the wier evokes the collective social presence at the weir, where free-flowing river water becomes controlled irrigation water (56:128).

It would appear, then, that preindustrial sociotechnical systems have much

in common with today' s machine-based technological systems: Both rely ex­

tensively on nonverbal cognition, both show enormous complexity and elabo­

ration, and both seem to generate reified notions rather than "authentic"

self-awareness. Moreover, the conditions of freedom in preindustrial societies

are falsely represented by focusing on the allegedly nonconstraining nature of

tool (as opposed to machine) use. Any sociotechnical system, ancient or mod­

em, primitive or industrialized, stems from the efforts of system builders who

attempt to create a network capable of resisting dissociation. As previously

argued, the use of ritual to coordinate productive activity in preindustrial

sociotechnical systems amounts to a form of domination and control, even

under stateless conditions. One can suggest, in fact, that both modem devices and preindustrial systems of ritual coordination are machines, as Latour

(57 : 1 29) defines the term: "A machine, as the name implies, is first of all a

machination, a stratagem, a kind of cunning, where borrowed forces keep one

another in check so that none can fly apart from the group." Latour refers here

to the role that machines play in uniting the constituent elements of modern

sociotechnical systems: Machines tie the assembled forces to one another in a

sustainable network (see 57: 103 41\ for telling discussions of the diesel en­

gine, the Kodak camera, and the telephone). To argue thus is not to deny that

the rise of the machine has brought about important, if as yet incompletely

understood, alterations in human sociotechnical activity. It is to stress that the

Standard View, with its division of human history into the Age of the Tool and

the Age of the Machine, substantially overstates the political and subjective implications of the rise of machines (50: 174).

What can the sociotechnical systems concept tell us about another kind of

rupture, the kind produced when a modem industrial technology or artifact is

adopted by a "traditional" society? A variant of the Standard View, perfectly

expressed in the film 'The Gods Must Be Crazy," alleges that the world-wide

distribution of industrial artifacts will inevitably tear out the foundations of

"authentic" traditional cultures and draw all the peoples in the world within the

grip of consumer ideology. Implicit in this view is a strong version of techno­

logical determinism, the doctrine that because there is only one way to make or

use a material artifact, every culture that adopts it will be forced to develop the

same social and labor relations. Because social information is so crudely encoded in artifacts, however, it is extremely unlikely that a transferred artifact will succeed in bringing with it the ideological structure that produced it. For

example, Hebdige (41 ) shows how motor scooters were deliberately developed

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 5 1 1

in Italy to signify the feminine as opposed to the masculine motorbike: The motor was covered and quiet, the curves were soft and the shapes rounded, and

so on. In Britain in the 1960s, however, the motor scooter was adopted by Mods, male and female, for whom it signified a European ("soft") image, as against the Rockers, who appropriated the motorcycle to signify an American

("hard") image.

Thus the "recipient" (appropriating) culture can reinterpret the transferred

artifact as it sees fit. No less should be expected of people in so-called "tradi­tional societies." According to the sociotechnical systems model, no such

thing as a "traditional society" exists. Every human society is a world in the

process of becoming, in which people are engaged in the active technological

elaboration, appropriation, and modification of artifacts as the means of com­ing to know themselves and of coordinating labor to sustain their lives. New

resources are unlikely to be ignored if they can be woven into an existing or

new activity system. An artifact' s determinative implications for labor in one context may be nullified if it is adopted to fulfill an essentially expressive function, as is the case for many "showpiece" industrial installations in Third W orId countries.

In a recent important essay, Schaniel (89) has stressed that the adoption of artifacts does not necessarily imply the adoption of the system of logic that produced the technology. Schaniel illustrates this point by discussing the his­

tory of Maori appropriation of iron artifacts. In the first phase, the Maori

ignored the artifacts, seeing little or no value in them. After some experimenta­tion, the Maori found that hoes and spades could be worked into their indige­

nous system of agriculture. European observers were shocked to find that the

Maori bound their hoes to short handles and used this implement from the squatting position. The favorite implement for levering up the ground re­mained the digging stick. The Maori later modified the digging stick by affixing to it a short piece of straight iron (89:496). Schaniel concludes that "the process of adopting and adapting introduced technology . . . does not imply that introduced technology does not lead to change, but the change is not pre-ordained by the technology adapted ... . The process of technological

adaptation is one where the introduced technology is adopted to the social processes of the adopting society, and not vice-versa" (89:496-98).

That said, the appropriation of modem technology, whether for productive

or symbolic purposes, may bring with it what Pelto calls "de-localization," the irreversible growth of dependence on nonlocal sources of energy (76: 1 66-68). As Pelto' s study of the snowmobile in Lapland suggests, de-localization may expand the geographical scope within which people actively appropriate arti­facts, with extensive implications for social and cultural change. It would be wrong, however, to try to predict the trajectory of such change from a tcchnical analysis of the transferred technology, as the extensive literature on the social

impact of the Green Revolution attests. According to some studies (36, 37), the Green Revolution invariably leads to "techno-economic differentiation" and

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

512 PFAFFENBERGER

the growth of a pauper class because rich farmers disproportionately benefit

from the extra-local resources (high-yielding varieties, pesticides, herbicides,

and fertilizers). Other studies report that Green Revolution technology does

not necessarily produce socioeconomic differentiation, so long as countervail­

ing customs assure the equitable use of agricultural inputs (3, 20, 3 1). In assessing the social and cultural impact of de-localization, however, it is

important to bear in mind that assuming technological determinism is much

easier than conducting a fully contextual study in which people are shown to

be the active appropriators, rather than the passive victims, of transferred

technology (79).

Sharp ' s famous analysis of steel axes among "stone-age" Australians illus­

trates the peril of reading too much technological determinism into a single case. Sharp showed how missionaries, by providing stone axes to women and

young men, whose status had previously been defined by having to ask tribal

elders for these artifacts, brought down a precariously legitimated stratification

system. However, any status differentiation system that depends on sumptuary

regulations, rules that deny certain artifacts to those deemed low in status, is

vulnerable to furious adjustment strategies if such artifacts suddenly become widely available; culture contact and technology transfer are by no means

required to set such processes in motion. The process Sharp described is not constitutive of technology transfer per se; a clear analogue is the erosion of the

medieval aristocracy' s status as peasants freed themselves from sumptuary regulations and acquired high-status artifacts (73: 135-36).

Where technological change has apparently disrupted so-called "traditional

societies," the villain is much more likely to be colonialism than technology. Colonialism disrupts indigenous political, legal, and ritual systems, and in so

doing, may seriously degrade the capacity of local system-builders to function

effectively within indigenous activity systems. In colonial Sri Lanka, the lib­

eral British government was obsessed with the eradication of multiple claims

to land, which were perceived to discourage investment and social progress. The legal eradication of such claims destroyed the ability of native headmen to adjust holdings to changing water supply levels and undermined the traditional

basis by which labor was coordinated for the repair of dams and irrigation canals. Village tanks and canals fell into disrepair as impecunious villagers allowed their lands to be taken over by village boutique owners and money­lenders (79). This example suggests that it is not transferred technology, but rather the imposition of an alien and hegemonic legal and political ideology­arguably, technicism, but not technology-that effects disastrous social

change in colonized countries.

It is when sociotechnical systems come into direct competition, as is the case in advanced technological diffusion, that spectacular disintegrations of

indigenous systems can occur. The sudden deployment of a competing system may outstrip the capacity of indigenous system participants to conceptualize their circumstances and make the necessary adjustments; their mode of de-

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 513

ploying resources, material and human, no longer works. Latour (58:32) com­ments:

The huge iron and steel plants of Lorraine are rusting away, no matter how many elements they tied together, because the world [their builders] were supposing to hold has changed. They are much like these beautiful words Scrabble players love to compose but which they do not know how to place on the board because the shape of the board has been modified by other players.

CONCLUSIONS

Against the Standard View' s exaggerated picture of technological evolution from simple tools to complex machines, the sociotechnical system concept puts forward a universal conception of human technological activity, in which complex social structures, nonverbal activity systems, advanced linguistic communication, the ritual coordination of labor, advanced artifact manufac­

ture, the linkage of phenomenally diverse social and nonsocial actors, and the social use of diverse artifacts are all recognized as parts of a single complex that is simultaneously adaptive and expressive.

The sociotechnical systems of the Machine Age do differ from their prein­dustrial predecessors, but the Standard View grossly exaggerates these differ­ences. For example, most modem definitions of technology assert that, unlike their preindustrial predecessors, modem technological systems are systems for the application of science, drawing their productive power from objective, linguistically encoded knowledge (e.g. 16). But on closer examination we see here the influence of Standard View mythology. Historians of technology tell us that virtually none of the technologies that structure our current social landscape were produced by the application of science; on the contrary, sci­

ence and organized objective knowledge are more commonly the result of technology. The principles of thermodynamics, for example, were discovered as scientists sought to determine how devices actually worked and what their operating parameters were (26). The notion that modem technology is efffec­tive because it is founded in objective , "true" knowledge violates the principle of symmetry advanced earlier in this essay, even as it denigrates the achieve­ments of preindustrial sociotechnical systems. As Lansing notes (56), Balinese water temples were more effective managers of irrigation than the all-but-dis-astrous Green Revolution techniques have been.

.

By jettisoning material-culture studies in the early 20th century, anthropol­ogy lost one means of developing a holistic, multi-disciplinary approach to culture. By reinstating the social anthropology of technology and material culture, we lay the foundation once again for fruitful communication among social anthropologists, ethnoarchaeologists, archaeologists, and students of human evolution. Besides challenging certain myths about technology that social anthropologists often take for granted, I hope this essay helps to raise

the level of such interdisciplinary discourse. For example, efforts are now

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

5 1 4 PFAFFENBERGER

underway to comprehend human evolution in terms of the complex interplay among "tools, language, and intelligence" (29). From the perspective of this essay, such an effort is misconceived: It overprivileges tools and language, and disguises the truly significant phenomena-namely, sociotechnical systems and nonverbal cognition. To grasp the evolutionary significance of human technological activity, I suggest that anthropologists lay aside the myths of the Standard View ("necessity is the mother of invention," "the meaning of an artifact is a surface matter of style," and "the history of technology is a unilinear progression from tools to machines"), and view human technological activity using the concept of the sociotechnical system. Once we do so, we can begin to construct hypotheses about the universals of human technology-uni­versals that highlight what is distinctly human about activities as diverse as making stone tools and launching space vehicles.

Literature Cited

1 . Ames, K. 1978. Meaning in artefacts: hall furnishings in Victorian America. 1. Inter­dis. Hist. 9: 1 9-46

2. Appadurai, A. 1986. Introduction: com­modites and the politics of value. In The Social Life of Things: Commodities in Cultural Perspective, ed. A. Appadurai, pp. 3-6 1 . Cambridge: Cambridge Univ. Press

3. Attwood, D. 1979. Why some of the poor get richer: economic change and mobility in rural Western India. Curro Anthropo!. 20:495-5 1 4

4. Basalla, G . 1 988. The Evolution of Tech­nology. Cambridge: Cambridge Univ. Press

5. BaudriJlard, J. 1975. The Mirror ofProduc­tion, transl. M. Poster. SI. Louis: Telos Press

6. Bijker, w., Hughes, T., Pinch, T., eds. 1987. The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology. Cambridge: MIT Press

7. Binford, L. 1 962. Archaeology as anthro­pology. Am. Antiq. 28:217-25

8. Binford, L. 1965. Archaeological systemat­ics and the stndy of cnltnre process. Am. Antiq. 3 1 :203-1 0

9 . Bloch, M . 1974. Symbol, song, dance, and the features of articnlation. Eur. J. Sociol. 1 5 :55-8 1 .

10. Bloch, M . 1 99 1 . Language, anthropology, and cognitive science. Man 26: 1 83-98

1 1 . Bloor, D. 199 1 . Knowledge and Social Im­agery. Chicago: Univ. Chicago Press. 2nd. ed.

12. Brown, R. 1987. Society as Text: Essays on Rhetoric, Reason, and Reality. Chicago: Univ. Chicago Press

1 3 . Bouquet, M. 199 1 . Images of artefacts. Crit. Anthropol. 1 1 :333-56

14. Bourdieu, P. 1 984. Distinction: A social

Critique of the Judgement of Taste. Lon­don: Routledge and Kegan Paul.

1 5. Bradbury, M., McFarlane, J. 1976. The name and nature of modernism. In Modern­ism, ed. M. Bradbury, 1. McFarlane, pp. 19-55. Harmondsworth: Penguin

16. Bruzina, R. 1 982. Art and architecture, an­cient and modern. Res. Phi/os. Techno!. 5: 1 63-87

17 . Burke, P. 1978. Popular Culture in Early Modern Europe. London: Temple Smith

1 8. Clifford, J., Marcus, G., eds. 1 986. Writing Culture: The Poetics and Politics of Eth­nography. Berkeley: Univ. Calif. Press.

1 9. Condominas, G. 1 986. Ritual technology in swidden agriculture. In Rice Societies: Asian Problems and Prospects, ed. I. Nor­lund, S. Cederroth, I. Gerden. pp. 29-37. Riverdale: Curzon Press

20. Coward, E. 1979. Principles of social or­ganization in an indigenous irrigation sys­tem. Hum. Organ. 38:35-43

2 1 . Digard, J . ..p. 1 979. La technologie en an­thropologie: fin de parcours ou nouveau souffle? L'Homme 1 9 : 105-40

22. Douglas, M., Isherwood, B. 1 979. The World of Goods. New York: Basic Books

23. Dunnell, R. 1 978. Style and fnnction: a fundamental dichotomy. Am. Antiq. 43: 1 92-202

24. Flannery , K. V., Marcus,J. 1 976. Formative Oaxaca and the Zapotec cosmos. Am. Sci. 64:374-83

25. Ferguson, E. S. 1 977. The mind's eye: non­verbal thought in technology. Science 197:827-36

26. Fores, M. 1 982. Technological change and the "technology" myth. Scand. Econ. Hist. Rev. 30: 1 67-88

27. Forty, A. 1 986. Objects of Desire. New York: Pantbeon

28. Gell, A. 1975. Metamorphosis of the Cas­sowaries. London: Athlonc Press

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

SOCIAL ANTHROPOLOGY OF TECHNOLOGY 5 1 5

29. Gibson, K. 1 99 1 . Tools, language, and in­telligence: evolutionary implications. Man 26:255--64

30. Giddens, A. 1979. Central Problems in So­cial Theory. London: MacMillan

3 1 . Goldman, H., Squire, L. 1 982. Technical change, labor use, and income distribution in the Muda irrigation project. Econ. Dev. Cult. Change 30:753-76

32. Golson, J. 1977. Simple 1001s and complex technology. In Stone Tools as Cultural Markers: Change, Evolution, and Com­plexity, ed. R. Wright, pp. 1 54- 1 6 1 . Can­berra: Australian Inst. Aborig. Stud.

33. Goody, E. 1978. Toward a theory of ques­tions. In Questions and Politeness: Strate­gies in Social Interaction, ed. E. Goody. Cambridge: Cambridge Univ. Press

34. Gould, R. 1968. Living archaeology: the Ngatatjara of Western Australia. Southwest. 1. Anthropol. 24: 1 0 1 -22

35. Granero, F. 1986. Power, ideology, and the ritual of production in lowland south Amer­ica. Man 2 1 :657-79

36. Griffin, K. 1 974. The Political Economy of Agrarian Change: An Essay on the Green Revolution. Cambridge: Harvard Univ. Press

37. Grolsch, K. 1 972. Technical change and Ihe destruction of income in rural areas. Am. 1. Agric. Ecoll. 54:326-41

38. Hally, D. 1985. The identification of vessel function: a case study from northwest Geor­gia. Am. Antiq. 5 1 :267-95

39. Harris, M. 1 968 . The Rise of Anthropologi­cal Theory. New York: Crowell

40. Harvey, D. 1 989. The Condition of Post­modernity: An Enquiry into the Origins of Cultural Change. Oxford: Blackwell

4 1 . Hebdige, D. 1 98 1 . Object as image: the Italian scooter cycle. Block 4:39-56

42. Heitzman, E. 1 987. Temple urbanism in medieval south India. J. Asian Stud. 46:791-826

43. Hodder, I. 1985. Reading the Past: Current Approaches to Interpretation in A r­chaeology. Cambridge: Cambridge Univ. Press

44. Hodder, I., ed. 1 989. The Meaning of Things: Malerial Cullure and Symbolic Ex­pression. London: Unwin Hyman

45. Hughes, T. 1983. Networks of Power: Elec­trification in Western Society, 1880-1930. Baltimore: Johns Hopkins Univ. Press

46. Hughes, T. 1987. The evolution of large technological systems. See Ref. 6, pp. 5 1 -82

47. Hughes, T. 1989. American Genesis: A Century of Innovation and Technological Enthusiasm, 1870-1970. New York: Pen­guin

48. Hughes, T. 1990. From deterministic dyna­mos to seamless-web systems. In Engineer­ing as a Social Enterprise, ed. H. Sladovich, pp. 7-25. Washington: Natl. Acad. Press

49. Hutton, J. H. 1 944. The place of material

culture in the study of anthropology. J. Roy. Anthropol. Inst. 74: 1-6

50. Ingold, T. 1988. Tools, minds, and ma­chines: an excursion into the philosophy of technology, Techniques et Cultures 12: 1 5 1 -76

5 1 . James, S. 1 979. Confections, concoctions, and conceptions. J. Anthropol. Soc. Oxford 10:83-95

52. Keller, C., Keller, J. 1 99 1 a. Thinking and Acting with Iron. Urbana: Beckman Inst.

53. Keller, C., Keller, J. 1 99 1b. Imaging in iron or thought is not inner speech. Presented at Wenner-Gren Found. Anthropol . Res. Symp. No. 1 1 2, Rethinking Linguistic Relativity

54. Kroeber, A. L. 1948. Anthropology. New York: Harcourt & Brace

55. Kroeber, A., Kluckhohn, C. 1952. Culture: A Critical Review of Concepts and Defini­tions. Cambridge: Harvard Univ. Press

56. Lansing, S. 199 1 . Priests and Program­mers: Technologies of Power in the Engi­neered Lanscape of Bali. Princeton: Princeton Univ. Press

57. Latour, B. 1987. Science in Action: How to Follow Scientists and Engineers Through Society. Cambridge: Harvard Univ. Press

58. Latour, B. 1986. The Prince for machines as well as for machinations. Paper read al the Seminar on Technology and Social Change, Univ. of Edinburgh (June 1 2-13, 1 986)

59. Lauer, P., ed. 1974. Readings in Material Culture. Brisbane: Queensland Univ. An­thropol. Mus.

60. Law, 1. 1987. Technology and heterogene­ous engineering: the case of Portuguese expansion. See Ref. 6, pp. 1 1 1-34

6 1 . Lee, R. 1979. The !Kung San: Men, Women, and Work in a Foraging Society. Cambridge: Cambridge Univ. Press

62. Lemonnier, P. 1986. The study of material culture today: toward an anthropology of technical systems. J. Anthropol. Archaeol. 5 : 1 47-86

63. Lemonnier, P. 1 989. Bark capes, arrow­heads, and Concorde: on social repre­sentations of technology. See Ref. 44, pp. 156-7 1

64. LeonI' ev, A. 1 98 1 . The problem of activity in psychology. See Ref. 1 02, pp. 37-7 1

65. Leroi-Gourhan, A. 1 943. Evolution et tech­niques. L 'homme et la matiere. Paris: Albin Michel

66. Leroi-Gourhan, A. 1 945. Evolution el tech­niques. Milieu et techniques. Paris: Albin Michel

67. Ludden, D. 1985. Peasant Society in South India. Princeton: Princeton Univ. Press

68. MacKenzie, D. 1 987. Missile accuracy: a case study in the social processes of tech­nological change. See Ref 6, pp. 195-222

69. Malinowski, B. 1935. Coral Gardens and Their Magic. London: Routledge

70. Marcus, G., Fischer, M. 1986. Anthropol-

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.

516 PFAFFENBERGER

ogy as Cultural Critique: An Experimental Moment in the Human Sciences. Chicago: Univ. Chicago Press.

7 1 . Mauss, M. 1983. Sociologie et anthropolo­gie. Paris: Presses Universitaires de France

72. Miller, D., ed. 1 983. Things ain't what they used to be." Roy. Anthropol. Inst. News 59:5-1 6

73. Miller, D . 1 987. Material Culture and Mass Consumption. London: Basil Black­well

74. Mulkay, M. 1 978. Science and the Sociol­ogy of Knowledge. London: George Allen and Unwin

75. Norman, D. 1988. The Psychology of Eve­ryday Things. New York: Basic Books

76. Pelto, P. 1 973. The Snowmobile Revolu­tion: Technology and Social Change in the Arctic. Menlo Park, CA: Benjamin Cum­mings

77. Pfaffenberger, B . 1 988. Festishized objects and humanized nature: toward an anthro­pology of technology. Man 23:236-52

78. Pfaffenberger, B. 1988. The social meaning of the personal computer, or, why the per­sonal computer revolution was no revolu­tion. Anthropol. Q. 6 1 :39-47

79. pfaffenberger, B. 1 990. The harsh facts of hydraulics: technology and society in Sri Lanka's colonization schemes. Technol. Cult. 3 1 : 361-97

80. Pfaffenberger, B. 1990. The Hindu temple as a machine, or the Western machine as a temple. Techniques et Cultures ( 1 6): 1 83-202

8 1 . Pfaffenberger, B. 1 990. Democratizing In­formation: Online Databases and the Rise of End- User Searching. Boston: G. K. Hall

82. Pfaffenberger, B. 1 992. Technological dra­mas. Sci., Technol., Human Values 17. In press

83. Rappaport, R. A. 1 97 1 . Ritual, sanctity, and cybernetics. Am. Anthropol. 73:59-76

84. Reynolds, B. 1 983. The relevance of mate­rial culture to anthropology. J. Anthropol. Soc. Oxford 14:209-17

85. Richardson, M. , ed. 1 974. The Human Mir­ror: Material and Spatial Images of Man. Baton Rouge: Louisiana State Univ. Press

86. Sahlins, M. 1 972. Stone Age Economics. Chicago: Aldine

87. Sahlins, M. 1 976. Culture and Practical Reason. Chicago: Univ. Chicago Press

88. Sayce, R. 1 933. Primitive Arts and Crafts: An Introduction to the Study of Material Culture. Cambridge: Cambridge Univ. Press

89. Schaniel, W. 1988. New technology and cultural change in traditional societies. J. Econ. Issues 22:493-98

90. Scheffler, I. 1967. Science and Subjectiv­ity. New York: Knopf

9 1 . Shanks, M., Tilley, C. 1987. Social Theory and Archaeology. Albuquerque: Univ. New Mexico Press

92. Sillitoe, P. 1 988. Made in Niugini: Technol­ogy in the Highlands of Papua New Guinea. London: Brit. Mus.

93. Spier, R. 1970. From the Hand of Man: Primitive and Preindustrial Technologies. Boston: Houghton-Mifflin

94. Staudenmaier, J. 1989. The politics of suc­cessful technologies. In In Context: His­tory and the History of Technology, ed. H. Cutliffe, R. Post, pp. 150-7 1 . Bethlehem: Lehigh Univ. Press

95. Tennekoon, S. 1 988. Rituals of develop­ment: the accelerated Mahaweli develop­ment program of Sri Lanka. Am. Ethnol. 1 5 :294--3 1 0

96. Tilley, C . 1990. Reading Material Culture. Oxford: Blackwell

97. Turner, Y. 1 957. Schism and Continuity in an African Society: A Study of Ndembu Village Life. Manchester: Manchester Univ. Press

98. Turner, Y. 1974. Dramas, Fields, and Metaphors: Symbolic Action in Human So­ciety. Ithaca: Cornell Univ. Press

99. Vicente, W. 1984. Technological knowl­edge without science: the innovation of flush riveting in American airplanes, ca. 1930-1950. Technol. Cult. 25:540-76.

100. Vygotsky, L. 1978. Mind in Society. Cam­bridge: Harvard Univ. Press

l U I . Wallace, A. 1 978. Rockdale: The Growth of an American Village in the Early Indus­trial Revolution. New York: Alfred A. Knopf

102. Wertsch, J., ed. 198 1 . The Concept of Ac­tivity in Soviet Psychology. Armonk, NY: M. E. Sharpe

1 03 . Wissler, C. 19 14. Material cultures of the North American Indians. Am. Anthropol. 1 6:447-505

Ann

u. R

ev. A

nthr

opol

. 199

2.21

:491

-516

. Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

IN

STIT

UT

O V

EN

EZ

OL

AN

O D

E I

NV

EST

IGA

CIO

NE

S C

IEN

TIF

ICA

S on

05/

03/1

0. F

or p

erso

nal u

se o

nly.